Treebank Statistics: UD_Maghrebi_Arabic_French-Arabizi: POS Tags: PROPN
There are 549 PROPN lemmas (10%), 786 PROPN types (10%) and 1866 PROPN tokens (9%).
Out of 16 observed tags, the rank of PROPN is: 3 in number of lemmas, 3 in number of types and 5 in number of tokens.
The 10 most frequent PROPN lemmas: Dieu, dieu_mien, Algérie, Saadane, verts, ziani, maroc, France, Algerie, Setif
The 10 most frequent PROPN types: rabi, allah, algerie, lah, khadra, ziani, sa3dan, alah, algérie, maroc
The 10 most frequent ambiguous lemmas: Dieu (PROPN 256, NOUN 1), dieu_mien (PROPN 4, NOUN 1), Algérie (PROPN 178, ADJ 1, NOUN 1), Cheikh (PROPN 9, NOUN 3), ben (INTJ 1, PROPN 1), Annaba (PROPN 6, PART 1), Hadj (PROPN 5, NOUN 2), Yebda (PROPN 4, VERB 1), fennecs (PROPN 3, NOUN 1), ramadan (PROPN 2, NOUN 1)
The 10 most frequent ambiguous types: rabi (PROPN 190, NOUN 3), allah (PROPN 91, X 9), lah (PROPN 26, NOUN 2, PART 1, PRON 1, X 1), alah (PROPN 19, X 1), ben (PROPN 11, NOUN 1), llah (PROPN 7, X 1), karim (PROPN 5, ADJ 1), chikh (PROPN 4, NOUN 1), algeria (PROPN 3, NOUN 1), cheikh (PROPN 3, NOUN 2)
- rabi
- allah
- lah
- PROPN 26: je suis avec eux lah yl3an had el khabith
- NOUN 2: makan lah telomni yana lom denia li rahi t3ayef
- PART 1: yekhi hleft 3ala 3000000 da lah mameditech azayed w medit moins khatrakech mechhah ya serrar
- PRON 1: lah yej3al el 7afd w l7amdoulah li makanche li matou
- X 1: je souhaite la reussite pour notre equipe inchaa lah et l’ égypte aussi comme ça on finit cette querelle nous sommes tous freres bonne chance à tous rabi 3awenkom
- alah
- ben
- llah
- karim
- chikh
- algeria
- cheikh
Morphology
The form / lemma ratio of PROPN is 1.431694 (the average of all parts of speech is 1.474223).
The 1st highest number of forms (44) was observed with the lemma “Dieu”: Allh, alaah, alah, alahe, alahh, alaho, alahouma, alhe, allah, allahe, allahi, allaho, allahoma, allahooma, allahou, allahouma, allahoumma, allahu, allho, alllah, dieu, elah, ellah, ellaho, ellahoima, ellahoma, elle, ilah, la, lah, lahhe, lahi, lahou, llah, llahi, llahou, lleh, rab, raba, rabe, rabou, rebbi, élah, الله.
The 2nd highest number of forms (41) was observed with the lemma “Algérie”: ‘Algérie, agarie, agerie, alegerie, alg, alge, algeire, algeres, algeri, algeria, algeriaaaa, algerie, algerieee, algeriè, algirie, algèrie, algérie, aljazaeer, djayaire, djazair, djazairia, djazayare, djazayer, djazire, djzair, dzair, dzaire, dzayer, jazaeer, jazair, jazayar, jazayare, jazayer, jazeer, jazeir, jzayer, lagerie, tzayer, zair, âlgerie, الجزائر.
The 3rd highest number of forms (13) was observed with the lemma “verts”: 5adra, kadera, khadhra, khadra, khadraaaaa, khedra, khoder, ver, veres, vert, vertes, verts, vets.
PROPN occurs with 2 features: Typo (5; 0% instances), AdpType (1; 0% instances)
PROPN occurs with 2 feature-value pairs: AdpType=Prep, Typo=Yes
PROPN occurs with 3 feature combinations.
The most frequent feature combination is _ (1860 tokens).
Examples: rabi, allah, algerie, lah, khadra, ziani, sa3dan, alah, algérie, maroc
Relations
PROPN nodes are attached to their parents using 16 different relations: nsubj (553; 30% instances), nmod (311; 17% instances), obj (205; 11% instances), obl (173; 9% instances), vocative (171; 9% instances), parataxis (149; 8% instances), flat (120; 6% instances), conj (82; 4% instances), appos (44; 2% instances), root (39; 2% instances), discourse (5; 0% instances), dep (4; 0% instances), dislocated (4; 0% instances), iobj (3; 0% instances), amod (2; 0% instances), ccomp (1; 0% instances)
Parents of PROPN nodes belong to 10 different parts of speech: VERB (1136; 61% instances), NOUN (302; 16% instances), PROPN (231; 12% instances), PRON (109; 6% instances), (39; 2% instances), ADJ (31; 2% instances), NUM (8; 0% instances), INTJ (6; 0% instances), ADP (3; 0% instances), ADV (1; 0% instances)
682 (37%) PROPN nodes are leaves.
735 (39%) PROPN nodes have one child.
337 (18%) PROPN nodes have two children.
112 (6%) PROPN nodes have three or more children.
The highest child degree of a PROPN node is 7.
Children of PROPN nodes are attached using 24 different relations: case (374; 21% instances), det (331; 18% instances), discourse (268; 15% instances), nmod (135; 7% instances), flat (116; 6% instances), cc (113; 6% instances), amod (93; 5% instances), conj (93; 5% instances), advmod (88; 5% instances), nsubj (65; 4% instances), parataxis (35; 2% instances), appos (20; 1% instances), punct (17; 1% instances), acl:relcl (14; 1% instances), mark (14; 1% instances), vocative (9; 0% instances), goeswith (4; 0% instances), nummod (3; 0% instances), compound (2; 0% instances), dep (2; 0% instances), obj (2; 0% instances), obl (2; 0% instances), acl (1; 0% instances), dislocated (1; 0% instances)
Children of PROPN nodes belong to 15 different parts of speech: ADP (371; 21% instances), DET (323; 18% instances), INTJ (267; 15% instances), PROPN (231; 13% instances), NOUN (130; 7% instances), CCONJ (112; 6% instances), ADJ (97; 5% instances), PRON (90; 5% instances), ADV (54; 3% instances), PART (41; 2% instances), VERB (32; 2% instances), SCONJ (18; 1% instances), PUNCT (17; 1% instances), NUM (12; 1% instances), X (7; 0% instances)